Seed implantation brachytherapy for prostate cancer has been projected to grow by 18-fold in the ten years since 1995, to become the preferred treatment option in 2005. The major obstacle to achieving uniformly high standard and cost-effective care in this procedure is the lack of robust and optimized treatment planning and real-time dosimetric guidance. The specific aim and long term objective of this proposed project are autonomous treatment planning and interventional treatment guidance, respectively. The core technologies include a genetic algorithm engine which designs the optimal seed implant for each patient, a decision- theoretic steering agent which ensures that the clinical objectives and tradeoffs are thoroughly explored in 10(2) second time scale, and a sensitivity analysis module which simulates and reduces the likelihood of real-time surgical deviations from the optimized plan. The maturity of web-based computing technology has enabled integration of these intelligent modules to create a synergy for immersive treatment optimization in both telemedical and intra-operative settings. The proposed planning and guidance system will be developed and integrated on the platform of these technologies. The system is targeted to eliminate redundant and error-prone steps, to streamline the clinician- machine interface, and to provide stepwise feedforward optimization. PROPOSED COMMERCIAL APPLICATION: The immediate commercial application will focus on the autonomous Internet or intranet based treatment planning and the real-time interventional intra-operative treatment guidance for prostate seed implantation. The scope of application may extend to radiation treatment planning in general for the telemedical network-based system. The real- time interventional systems have strong potential in the treatment of restenosis and in the setting of cath-lab.